Anthramycin is a member of a class of natural compounds named pyrrolo[1,4]benzodiazepines (PDBs) or, more simply, the benzodiazepine antibiotics. Members of the benzodiazepine antibiotics include the compounds sibiromycin, tomaymycin, neothramycin, porothramycin, sibanomycin, mazethramycin, DC-81, chicamycin and abbeymycin. Naturally occurring benzodiazepine antibiotics are structurally related tricyclic compounds, consisting of an aromatic-ring, a 1,4-diazepin-5-one-ring bearing a N10-C11 imine-carbinolamine moiety, and a pyrrol-ring, as shown below. Different patterns of substitution of the three rings distinguish the different members of this antibiotic class.

Precursor feeding studies have established the biosynthetic building blocks for anthramycin (Hurley et al., 1975, J. Am. Chem. Soc., 97(15), 4372–4378). The anthranilate moieties of these antibiotics are derived from tryptophan via the kynurenine pathway, with the three antibiotics differing in the pattern of substitution at the aromatic ring (Hurley & Gariola, 1979 Antimicrob. Agents Chemother. 15:42–45). The 2-carbon and 3-carbon proline units of the antibiotics are derived from catabolism of L-tyrosine. The additional carbon atom found in the 3-carbon proline unit of anthramycin and sibiromycin is derived from methionine and is absent in the 2-carbon proline unit of tomaymycin. Despite the precursor feeding studies, the genes and proteins forming the biosynthetic locus for producing anthramycin have remained unidentified. Many low molecular weight peptides produced by bacteria are synthesized nonribosomally on large multifunctional proteins termed peptide synthetases (Konz & Marahiel, 1999, Chem. Biol., Vol. 6, pp. R39–R48). Nonribosomal peptide synthetases (NRPSs) are multidomain proteins that contain sets of functional domains arranged into units called modules. Konz & Marahiel (1999) teaches that the sinqie domains can be detected easily in sequence aliqnments by the presence of highly conserved signature or ‘core’ sequences that impart each domain type a characteristic fingerprint, irrespective of the enzyme origin, and discloses the conserved motifs of functional domains (FIG. 3 and Table 2 of Konz & Marahiel).
Benzodiazepine antibiotics have been shown to possess potent biological activitities, including antibiotic, antitumor and antiviral activities (Hurley, 1977, J. Antibiot. 30:349). However, clinical use of benzodiazepine has been compromised primarily because of dose-limiting cardiotoxicity. Consequently, considerable effort has been devoted to creating heterocyclic analogs of the benzodiazepine antibiotics that would retain the desired antitumor activities while avoiding the formation of cardiotoxic quinone-amine products. Elucidation of gene clusters involved in the biosynthesis of benzodiazepines expands the repertoire of genes and proteins useful to produce benzodiazepines via combinatorial biosynthesis.
There is great interest in discovering and developing small molecules capable of binding to DNA in a sequence-selective manner. Anthramycin binds the minor groove of DNA and generates covalent adducts at the 2-amino group of guanine bases. Anthramycin minor groove binding exhibits G-C base specificity. The sequence A-G-A is most favored of all, perhaps because it allows drug binding in either orientation (the acrylamide tail binds at the 5′ position of the binding site and prefers the deep minor groove of an AT pair; G-G-G is disfavored because it makes no accommodation for the acrylamide tail in either direction). Compounds having the potential to target and down-regulate individual genes would be useful in the therapy of genetic-based diseases such as cancer. Such compounds would also be useful in diagnostics, functional genomics and target validation (Thurston et al. 1999, J. Med. Chem. 42:1951–1964). Elucidation of the genes and proteins forming the biosynthetic locus for anthramycin provides a means of generating small molecules capable of binding to DNA in a sequence selective manner.
Existing screening methods for identifying benzodiazepine-producing microbes are laborious, time consuming and have not provided sufficient discrimination to date to detect organisms producing benzodiazepine natural products at low levels. There is a need for tools capable of detecting organisms that produce benzodiazepines at levels that are not detected by traditional culture tests.